The relationship between prevalence of Hantavirus (primarily Sin Nombre Virus, SNV) infection and population density will be evaluated in small mammal carriers in the southwestern United States. In addition, we will develop a model using Geographic Information System (GIS) technology to predict environmental correlates of high human hantaviral risk. A combination of descriptive and analytic studies of environmental characteristics associated with rodent Hantavirus prevalence will be used in the study in conjunction with field manipulation experiments of closed small mammal populations to study these problems. The major hypotheses we intend to test ar that (1) rodent population density is positively associated with incidence of rodent Hantavirus infection, and (2) environmental characteristics, such as rainfall and vegetation, (which can be modelled and identified across geographic space using GIS technology), are associated with the level of Hantavirus infection in rodent populations. Results of these studies will provide information on the degree of human risk associated with the virus and allow predictions regarding future outbreaks.